Date:	Tue, 19 Apr 2016 13:53:17 -0400
From:	Jeff Layton <jlayton@...chiereds.net>
To:	Al Viro <viro@...IV.linux.org.uk>, linux-cifs@...r.kernel.org
Cc:	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/6] cifs: quit playing games with draining iovecs

On Sat, 2016-04-09 at 21:51 +0100, Al Viro wrote:
> ... and use ITER_BVEC for the page part of request to send
> 
> Signed-off-by: Al Viro <viro@...iv.linux.org.uk>
> ---
>  fs/cifs/cifsproto.h |   2 -
>  fs/cifs/transport.c | 141 +++++++++++++++---------------------------
> ----------
>  2 files changed, 41 insertions(+), 102 deletions(-)
> 
> diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h
> index d9b4f44..7d5f53a 100644
> --- a/fs/cifs/cifsproto.h
> +++ b/fs/cifs/cifsproto.h
> @@ -37,8 +37,6 @@ extern void cifs_buf_release(void *);
>  extern struct smb_hdr *cifs_small_buf_get(void);
>  extern void cifs_small_buf_release(void *);
>  extern void free_rsp_buf(int, void *);
> -extern void cifs_rqst_page_to_kvec(struct smb_rqst *rqst, unsigned
> int idx,
> -					struct kvec *iov);
>  extern int smb_send(struct TCP_Server_Info *, struct smb_hdr *,
>  			unsigned int /* length */);
>  extern unsigned int _get_xid(void);
> diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c
> index 87abe8e..206a597 100644
> --- a/fs/cifs/transport.c
> +++ b/fs/cifs/transport.c
> @@ -124,41 +124,32 @@ cifs_delete_mid(struct mid_q_entry *mid)
>  /*
>   * smb_send_kvec - send an array of kvecs to the server
>   * @server:	Server to send the data to
> - * @iov:	Pointer to array of kvecs
> - * @n_vec:	length of kvec array
> + * @smb_msg:	Message to send
>   * @sent:	amount of data sent on socket is stored here
>   *
>   * Our basic "send data to server" function. Should be called with
> srv_mutex
>   * held. The caller is responsible for handling the results.
>   */
>  static int
> -smb_send_kvec(struct TCP_Server_Info *server, struct kvec *iov,
> size_t n_vec,
> -		size_t *sent)
> +smb_send_kvec(struct TCP_Server_Info *server, struct msghdr
> *smb_msg,
> +	      size_t *sent)
>  {
>  	int rc = 0;
> -	int i = 0;
> -	struct msghdr smb_msg;
> -	unsigned int remaining;
> -	size_t first_vec = 0;
> +	int retries = 0;
>  	struct socket *ssocket = server->ssocket;
>  
>  	*sent = 0;
>  
> -	smb_msg.msg_name = (struct sockaddr *) &server->dstaddr;
> -	smb_msg.msg_namelen = sizeof(struct sockaddr);
> -	smb_msg.msg_control = NULL;
> -	smb_msg.msg_controllen = 0;
> +	smb_msg->msg_name = (struct sockaddr *) &server->dstaddr;
> +	smb_msg->msg_namelen = sizeof(struct sockaddr);
> +	smb_msg->msg_control = NULL;
> +	smb_msg->msg_controllen = 0;
>  	if (server->noblocksnd)
> -		smb_msg.msg_flags = MSG_DONTWAIT + MSG_NOSIGNAL;
> +		smb_msg->msg_flags = MSG_DONTWAIT + MSG_NOSIGNAL;
>  	else
> -		smb_msg.msg_flags = MSG_NOSIGNAL;
> -
> -	remaining = 0;
> -	for (i = 0; i < n_vec; i++)
> -		remaining += iov[i].iov_len;
> +		smb_msg->msg_flags = MSG_NOSIGNAL;
>  
> -	i = 0;
> -	while (remaining) {
> +	while (msg_data_left(smb_msg)) {
>  		/*
>  		 * If blocking send, we try 3 times, since each can
> block
>  		 * for 5 seconds. For nonblocking  we have to try
> more
> @@ -177,35 +168,21 @@ smb_send_kvec(struct TCP_Server_Info *server,
> struct kvec *iov, size_t n_vec,
>  		 * after the retries we will kill the socket and
>  		 * reconnect which may clear the network problem.
>  		 */
> -		rc = kernel_sendmsg(ssocket, &smb_msg,
> &iov[first_vec],
> -				    n_vec - first_vec, remaining);
> +		rc = sock_sendmsg(ssocket, smb_msg);
>  		if (rc == -EAGAIN) {
> -			i++;
> -			if (i >= 14 || (!server->noblocksnd && (i >
> 2))) {
> +			retries++;
> +			if (retries >= 14 ||
> +			    (!server->noblocksnd && (retries > 2)))
> {
>  				cifs_dbg(VFS, "sends on sock %p
> stuck for 15 seconds\n",
>  					 ssocket);
> -				rc = -EAGAIN;
> -				break;
> +				return -EAGAIN;
>  			}
> -			msleep(1 << i);
> +			msleep(1 << retries);
>  			continue;
>  		}
>  
>  		if (rc < 0)
> -			break;
> -
> -		/* send was at least partially successful */
> -		*sent += rc;
> -
> -		if (rc == remaining) {
> -			remaining = 0;
> -			break;
> -		}
> -
> -		if (rc > remaining) {
> -			cifs_dbg(VFS, "sent %d requested %d\n", rc,
> remaining);
> -			break;
> -		}
> +			return rc;
>  
>  		if (rc == 0) {
>  			/* should never happen, letting socket clear
> before
> @@ -215,59 +192,11 @@ smb_send_kvec(struct TCP_Server_Info *server,
> struct kvec *iov, size_t n_vec,
>  			continue;
>  		}
>  
> -		remaining -= rc;
> -
> -		/* the line below resets i */
> -		for (i = first_vec; i < n_vec; i++) {
> -			if (iov[i].iov_len) {
> -				if (rc > iov[i].iov_len) {
> -					rc -= iov[i].iov_len;
> -					iov[i].iov_len = 0;
> -				} else {
> -					iov[i].iov_base += rc;
> -					iov[i].iov_len -= rc;
> -					first_vec = i;
> -					break;
> -				}
> -			}
> -		}
> -
> -		i = 0; /* in case we get ENOSPC on the next send */
> -		rc = 0;
> +		/* send was at least partially successful */
> +		*sent += rc;
> +		retries = 0; /* in case we get ENOSPC on the next
> send */
>  	}
> -	return rc;
> -}
> -
> -/**
> - * rqst_page_to_kvec - Turn a slot in the smb_rqst page array into a
> kvec
> - * @rqst: pointer to smb_rqst
> - * @idx: index into the array of the page
> - * @iov: pointer to struct kvec that will hold the result
> - *
> - * Helper function to convert a slot in the rqst->rq_pages array
> into a kvec.
> - * The page will be kmapped and the address placed into iov_base.
> The length
> - * will then be adjusted according to the ptailoff.
> - */
> -void
> -cifs_rqst_page_to_kvec(struct smb_rqst *rqst, unsigned int idx,
> -			struct kvec *iov)
> -{
> -	/*
> -	 * FIXME: We could avoid this kmap altogether if we used
> -	 * kernel_sendpage instead of kernel_sendmsg. That will only
> -	 * work if signing is disabled though as sendpage inlines
> the
> -	 * page directly into the fraglist. If userspace modifies
> the
> -	 * page after we calculate the signature, then the server
> will
> -	 * reject it and may break the connection. kernel_sendmsg
> does
> -	 * an extra copy of the data and avoids that issue.
> -	 */
> -	iov->iov_base = kmap(rqst->rq_pages[idx]);
> -
> -	/* if last page, don't send beyond this offset into page */
> -	if (idx == (rqst->rq_npages - 1))
> -		iov->iov_len = rqst->rq_tailsz;
> -	else
> -		iov->iov_len = rqst->rq_pagesz;
> +	return 0;
>  }
>  
>  static unsigned long
> @@ -299,8 +228,9 @@ smb_send_rqst(struct TCP_Server_Info *server,
> struct smb_rqst *rqst)
>  	unsigned int smb_buf_length =
> get_rfc1002_length(iov[0].iov_base);
>  	unsigned long send_length;
>  	unsigned int i;
> -	size_t total_len = 0, sent;
> +	size_t total_len = 0, sent, size;
>  	struct socket *ssocket = server->ssocket;
> +	struct msghdr smb_msg;
>  	int val = 1;
>  
>  	if (ssocket == NULL)
> @@ -321,7 +251,13 @@ smb_send_rqst(struct TCP_Server_Info *server,
> struct smb_rqst *rqst)
>  	kernel_setsockopt(ssocket, SOL_TCP, TCP_CORK,
>  				(char *)&val, sizeof(val));
>  
> -	rc = smb_send_kvec(server, iov, n_vec, &sent);
> +	size = 0;
> +	for (i = 0; i < n_vec; i++)
> +		size += iov[i].iov_len;
> +
> +	iov_iter_kvec(&smb_msg.msg_iter, WRITE | ITER_KVEC, iov,
> n_vec, size);
> +
> +	rc = smb_send_kvec(server, &smb_msg, &sent);
>  	if (rc < 0)
>  		goto uncork;
>  
> @@ -329,11 +265,16 @@ smb_send_rqst(struct TCP_Server_Info *server,
> struct smb_rqst *rqst)
>  
>  	/* now walk the page array and send each page in it */
>  	for (i = 0; i < rqst->rq_npages; i++) {
> -		struct kvec p_iov;
> -
> -		cifs_rqst_page_to_kvec(rqst, i, &p_iov);
> -		rc = smb_send_kvec(server, &p_iov, 1, &sent);
> -		kunmap(rqst->rq_pages[i]);
> +		size_t len = i == rqst->rq_npages - 1
> +				? rqst->rq_tailsz
> +				: rqst->rq_pagesz;
> +		struct bio_vec bvec = {
> +			.bv_page = rqst->rq_pages[i],
> +			.bv_len = len
> +		};
> +		iov_iter_bvec(&smb_msg.msg_iter, WRITE | ITER_BVEC,
> +			      &bvec, 1, len);
> +		rc = smb_send_kvec(server, &smb_msg, &sent);
>  		if (rc < 0)
>  			break;
>  

What's the advantage of using iov_iter_bvec over iov_iter_kvec ?

That said...

Acked-by: Jeff Layton <jlayton@...chiereds.net>

Powered by blists - more mailing lists